Source separation and denoising for hyperspectral fiber photometry neuromodulator imaging datasets.
conda create -n sourcesep python=3.8
conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia
conda install scipy scikit-learn statsmodels jupyterlab pandas seaborn h5py pytables
pip install lightning tensorboard rich timebudget autopep8 pytest
pip install xarray hvplot
pip install -e .
Config.toml contents:
['all']
root='/data/'
spectra='/data/spectra/'
pilot='/data/pilot/'
Data directory structure. (Download data from this dropbox link)
data
├── pilot
│ ├── GCaMP8s_1.csv
│ ├── GCaMP8s_2.csv
│ ├── GCaMP8s_3.csv
│ └── test.hdf5
├── sims
│ ├── 2023-02-24.h5
│ └── 2023-02-24.toml
├── sim_config.toml
├── calibrate_px_to_nm.tif
└── spectra
├── EGFP.csv
├── HbAbs.csv
├── Venus.csv
├── mApple.csv
└── pathlength.csv
./qdocs contains source files for the quarto project rendered in ./docs, which can be viewed as a github pages website.
Rohan Gala, Smrithi Sunil, Kaspar Podgorski, Uygar Sümbül